JOURNAL OF SCHOOL OF ENGINEERING
Year:2002 | Volume:14 | Issue:1|Papers Count:9

In this paper, a method is presented for the solution of general frictionless nonlinear contact-impact problems called "United Elements". In the fields of mechanics and structures, contact-impact problems are considered one of the most difficult fields. The reason is that in addition to the presence of material and geometric nonlinearities, the last possible nonlinearity, the boundary condition nonlinearity, enters the problem. In fact, the boundary condition will be obtained as a part of the solution. The usual methods for these problems stem from Lagrange Multipliers and Penalty Function methods, and in each method, the simultaneous satisfaction of contact constraints and equilibrium equations is sought. Both of these methods have some intrinsic mathematical and numerical deficiencies which relates to the quality of boundary condition exertion. Therefore, in this work, bayed on the physical concepts and mechanics principals, another method is used for the solution of contact-impact problems. The method is" not only suitable for computer programming but also compensates the deficiencies of the other methods.

An experimental investigation has been performed to determine the affecting parameters on the steady state performance of a two phase closed thermosyphon with water as the working fluid. The experimental results show a negligible effect of the mass of the fluid in thd pipe on the maximum heat load of the thermosyphon. The minimum mass of the fluid has been determined. The pipe power variation with entering vapour temperatare of the condenser has been compared with Nusselt Theory. Other parameters determined from the experiments are: effect of condensing water and the minimum mass value at each power, optimum mass for maximum power, boiling and condensation heat transfer coefficient in evaporator and condenser temperature difference. The results of this experiment are useful in design and optimization of thermosyphon for water and other fluids.

The scope of this work is the production of activated carbon in the laboratory scale. The activated carbon is prepared from walnut shell which is chemically activated by zinc chloride. It is observed that as the ratio of activating agent to raw material increases to 100%, a sharp enhancement of surface area, pore volume and iodine number of final product (activated carbon) results. However, by increasing the ratio to 200%, only the surface area and pore volume is gradually enhanced and iodine number changes inconsistently. It is also concluded that the size of raw material granules, tested in this study, shows no considerable effect on the surface area and adsorption properties of the product.

This paper presents a learning algorithm that guarantees stability for a class of closed loop neural network control systems. The underlying control system, "Plant", represents a nonlinear system. The neuro-controller, which indeed represents a direct adaptive controller, guarantees the closed loop stability for any arbitrary initial values of states and any initial neural network weight matrices, provided that the adaptive rates of neuro-controller be high enough No pre-learning is required.

In this paper, a conventional controller structure is employed and an optimal mixed H2/H∞ strategy to control the roll channel of a missile is designed. It is apparent that if the conventional output feedback is used, the controller order will be high and hence not easily implementable for practical purpose. The genetic algorithm has been used since the optimal mixed H2/H∞ control with conventional structure has no closed form solution. Unlike the other method, this method may also consider uncertainties; therefore the controller would have an optimal performance in terms of tracking and disturbance attenuation. Simulation results show the desired robustness and disturbance attenuation properties and also efficiency of the proposed technique compared to the conventional approach.

In this paper, an adaptive fuzzy sliding mode (AFSM) controller is proposed for induction motor drives, which combines" the merits of sliding mode control the fuzzy inference mechanism and the adaptive algorithm. First, a sliding mode speed controller with and integral operation-switching surface is designed, in which the acceleration information for speed control is not required. The exponential stability is guaranteed and insensitivity to uncertainties and disturbances are obtained as well In the design of sliding mode controller, the bound of the uncertainties must be available. In this paper, the significant of this parameter in regard to the system performance is shown. Next, a fuzzy sliding mode speed controller investigated in which a fuzzy interface mechanism is used to estimate the upper bound of uncertainties. Finally an adaptive algorithm is proposed to determine the optimized centers of membership functions for fuzzy estimator. The proposed approaches have been used in position and speed control drives. Computer simulation demonstrating the validity of the proposed adaptive fuzzy algorithm is provided.

In this paper, some of the existing methods of approximation are presented. Then to reduce the computational time, a new method for approximation of the structural responses is outlined. For optimization, method of penalty function is used and a new penalty function is introduced for discrete optimization. Some space structures are optimized and results are compared with other methods.